Deep Learning-Based Computer-Aided Diagnosis of Osteochondritis Dissecans of the Humeral Capitellum Using Ultrasound Images.

BACKGROUND: Ultrasonography is used to diagnose osteochondritis dissecans (OCD) of the humerus; however, its reliability depends on the technical proficiency of the examiner. Recently, computer-aided diagnosis (CAD) using deep learning has been applied in the field of medical science, and high diagnostic accuracy has been reported. We aimed to develop a deep learning-based CAD system for OCD detection on ultrasound images and to evaluate the accuracy of OCD detection using the CAD system. METHODS: The CAD process comprises 2 steps: humeral capitellum detection using an object-detection algorithm and OCD classification using an image classification network. Four-directional ultrasound images of the elbow of the throwing arm of 196 baseball players (mean age, 11.2 years), including 104 players with normal findings and 92 with OCD, were used for training and validation. An external dataset of 20 baseball players (10 with normal findings and 10 with OCD) was used to evaluate the accuracy of the CAD system. A confusion matrix and the area under the receiver operating characteristic curve (AUC) were used to evaluate the system. RESULTS: Clinical evaluation using the external dataset resulted in high AUCs in all 4 directions: 0.969 for the anterior long axis, 0.966 for the anterior short axis, 0.996 for the posterior long axis, and 0.993 for the posterior short axis. The accuracy of OCD detection thus exceeded 0.9 in all 4 directions. CONCLUSIONS: We propose a deep learning-based CAD system to detect OCD lesions on ultrasound images. The CAD system achieved high accuracy in all 4 directions of the elbow. This CAD system with a deep learning model may be useful for OCD screening during medical checkups to reduce the probability of missing an OCD lesion. LEVEL OF EVIDENCE: Diagnostic Level II. See Instructions for Authors for a complete description of levels of evidence.

Enhancing fracture diagnosis in pelvic X-rays by deep convolutional neural network with synthesized images from 3D-CT.

Pelvic fractures pose significant challenges in medical diagnosis due to the complex structure of the pelvic bones. Timely diagnosis of pelvic fractures is critical to reduce complications and mortality rates. While computed tomography (CT) is highly accurate in detecting pelvic fractures, the initial diagnostic procedure usually involves pelvic X-rays (PXR). In recent years, many deep learning-based methods have been developed utilizing ImageNet-based transfer learning for diagnosing hip and pelvic fractures. However, the ImageNet dataset contains natural RGB images which are different than PXR. In this study, we proposed a two-step transfer learning approach that improved the diagnosis of pelvic fractures in PXR images. The first step involved training a deep convolutional neural network (DCNN) using synthesized PXR images derived from 3D-CT by digitally reconstructed radiographs (DRR). In the second step, the classification layers of the DCNN were fine-tuned using acquired PXR images. The performance of the proposed method was compared with the conventional ImageNet-based transfer learning method. Experimental results demonstrated that the proposed DRR-based method, using 20 synthesized PXR images for each CT, achieved superior performance with the area under the receiver operating characteristic curves (AUROCs) of 0.9327 and 0.8014 for visible and invisible fractures, respectively. The ImageNet-based method yields AUROCs of 0.8908 and 0.7308 for visible and invisible fractures, respectively.

Deep learning-based osteochondritis dissecans detection in ultrasound images with humeral capitellum localization.

PURPOSE: Osteochondritis dissecans (OCD) of the humeral capitellum is a common cause of elbow disorders, particularly among young throwing athletes. Conservative treatment is the preferred treatment for managing OCD, and early intervention significantly influences the possibility of complete disease resolution. The purpose of this study is to develop a deep learning-based classification model in ultrasound images for computer-aided diagnosis. METHODS: This paper proposes a deep learning-based OCD classification method in ultrasound images. The proposed method first detects the humeral capitellum detection using YOLO and then estimates the OCD probability of the detected region probability using VGG16. We hypothesis that the performance will be improved by eliminating unnecessary regions. To validate the performance of the proposed method, it was applied to 158 subjects (OCD: 67, Normal: 91) using five-fold-cross-validation. RESULTS: The study demonstrated that the humeral capitellum detection achieved a mean average precision (mAP) of over 0.95, while OCD probability estimation achieved an average accuracy of 0.890, precision of 0.888, recall of 0.927, F1 score of 0.894, and an area under the curve (AUC) of 0.962. On the other hand, when the classification model was constructed for the entire image, accuracy, precision, recall, F1 score, and AUC were 0.806, 0.806, 0.932, 0.843, and 0.928, respectively. The findings suggest the high-performance potential of the proposed model for OCD classification in ultrasonic images. CONCLUSION: This paper introduces a deep learning-based OCD classification method. The experimental results emphasize the effectiveness of focusing on the humeral capitellum for OCD classification in ultrasound images. Future work should involve evaluating the effectiveness of employing the proposed method by physicians during medical check-ups for OCD.

Teeth and prostheses detection in dental panoramic X-rays using CNN-based object detector and a priori knowledge-based algorithm.

Deep learning techniques for automatically detecting teeth in dental X-rays have gained popularity, providing valuable assistance to healthcare professionals. However, teeth detection in X-ray images is often hindered by alterations in tooth appearance caused by dental prostheses. To address this challenge, our paper proposes a novel method for teeth detection and numbering in dental panoramic X-rays, leveraging two separate CNN-based object detectors, namely YOLOv7, for detecting teeth and prostheses, alongside an optimization algorithm to refine the outcomes. The study utilizes a dataset of 3138 radiographs, of which 2553 images contain prostheses, to build a robust model. The tooth and prosthesis detection algorithms perform excellently, achieving mean average precisions of 0.982 and 0.983, respectively. Additionally, the trained tooth detection model is verified using an external dataset, and six-fold cross-validation is conducted to demonstrate the proposed method's feasibility and robustness. Moreover, the investigation of performance improvement resulting from the inclusion of prosthesis information in the teeth detection process reveals a marginal increase in the average F1-score, rising from 0.985 to 0.987 compared to the sole teeth detection method. The proposed method is unique in its approach to numbering teeth as it incorporates prosthesis information and considers complete restorations such as dental implants and dentures of fixed bridges during the teeth enumeration process, which follows the universal tooth numbering system. These advancements hold promise for automating dental charting processes.

A Machine Learning-Based Severity Prediction Tool for the Michigan Neuropathy Screening Instrument.

Diabetic sensorimotor polyneuropathy (DSPN) is a serious long-term complication of diabetes, which may lead to foot ulceration and amputation. Among the screening tools for DSPN, the Michigan neuropathy screening instrument (MNSI) is frequently deployed, but it lacks a straightforward rating of severity. A DSPN severity grading system has been built and simulated for the MNSI, utilizing longitudinal data captured over 19 years from the Epidemiology of Diabetes Interventions and Complications (EDIC) trial. Machine learning algorithms were used to establish the MNSI factors and patient outcomes to characterise the features with the best ability to detect DSPN severity. A nomogram based on multivariable logistic regression was designed, developed and validated. The extra tree model was applied to identify the top seven ranked MNSI features that identified DSPN, namely vibration perception (R), 10-gm filament, previous diabetic neuropathy, vibration perception (L), presence of callus, deformities and fissure. The nomogram's area under the curve (AUC) was 0.9421 and 0.946 for the internal and external datasets, respectively. The probability of DSPN was predicted from the nomogram and a DSPN severity grading system for MNSI was created using the probability score. An independent dataset was used to validate the model's performance. The patients were divided into four different severity levels, i.e., absent, mild, moderate, and severe, with cut-off values of 10.50, 12.70 and 15.00 for a DSPN probability of less than 50, 75 and 100%, respectively. We provide an easy-to-use, straightforward and reproducible approach to determine prognosis in patients with DSPN.

A nomogram-based diabetic sensorimotor polyneuropathy severity prediction using Michigan neuropathy screening instrumentations.

BACKGROUND: Diabetic Sensorimotor polyneuropathy (DSPN) is one of the major indelible complications in diabetic patients. Michigan neuropathy screening instrumentation (MNSI) is one of the most common screening techniques used for DSPN, however, it does not provide any direct severity grading system. METHOD: For designing and modeling the DSPN severity grading systems for MNSI, 19 years of data from Epidemiology of Diabetes Interventions and Complications (EDIC) clinical trials were used. Different Machine learning-based feature ranking techniques were investigated to identify the important MNSI features associated with DSPN diagnosis. A multivariable logistic regression-based nomogram was generated and validated for DSPN severity grading using the best performing top-ranked MNSI features. RESULTS: Top-10 ranked features from MNSI features: Appearance of Feet (R), Ankle Reflexes (R), Vibration perception (L), Vibration perception (R), Appearance of Feet (L), 10-gm filament (L), Ankle Reflexes (L), 10-gm filament (R), Bed Cover Touch, and Ulceration (R) were identified as important features for identifying DSPN by Multi-Tree Extreme Gradient Boost model. The nomogram-based prediction model exhibited an accuracy of 97.95% and 98.84% for the EDIC test set and an independent test set, respectively. A DSPN severity score technique was generated for MNSI from the DSPN severity prediction model. DSPN patients were stratified into four severity levels: absent, mild, moderate, and severe using the cut-off values of 17.6, 19.1, 20.5 for the DSPN probability less than 50%, 75%-90%, and above 90%, respectively. CONCLUSIONS: The findings of this work provide a machine learning-based MNSI severity grading system which has the potential to be used as a secondary decision support system by health professionals in clinical applications and large clinical trials to identify high-risk DSPN patients.

Detecting pelvic fracture on 3D-CT using deep convolutional neural networks with multi-orientated slab images.

Pelvic fracture is one of the leading causes of death in the elderly, carrying a high risk of death within 1 year of fracture. This study proposes an automated method to detect pelvic fractures on 3-dimensional computed tomography (3D-CT). Deep convolutional neural networks (DCNNs) have been used for lesion detection on 2D and 3D medical images. However, training a DCNN directly using 3D images is complicated, computationally costly, and requires large amounts of training data. We propose a method that evaluates multiple, 2D, real-time object detection systems (YOLOv3 models) in parallel, in which each YOLOv3 model is trained using differently orientated 2D slab images reconstructed from 3D-CT. We assume that an appropriate reconstruction orientation would exist to optimally characterize image features of bone fractures on 3D-CT. Multiple YOLOv3 models in parallel detect 2D fracture candidates in different orientations simultaneously. The 3D fracture region is then obtained by integrating the 2D fracture candidates. The proposed method was validated in 93 subjects with bone fractures. Area under the curve (AUC) was 0.824, with 0.805 recall and 0.907 precision. The AUC with a single orientation was 0.652. This method was then applied to 112 subjects without bone fractures to evaluate over-detection. The proposed method successfully detected no bone fractures in all except 4 non-fracture subjects (96.4%).

Optimization technique combined with deep learning method for teeth recognition in dental panoramic radiographs.

Computer-assisted analysis of dental radiograph in dentistry is getting increasing attention from the researchers in recent years. This is mainly because it can successfully reduce human-made error due to stress, fatigue or lack of experience. Furthermore, it reduces diagnosis time and thus, improves overall efficiency and accuracy of dental care system. An automatic teeth recognition model is proposed here using residual network-based faster R-CNN technique. The detection result obtained from faster R-CNN is further refined by using a candidate optimization technique that evaluates both positional relationship and confidence score of the candidates. It achieves 0.974 and 0.981 mAPs for ResNet-50 and ResNet-101, respectively with faster R-CNN technique. The optimization technique further improves the results i.e. F1 score improves from 0.978 to 0.982 for ResNet-101. These results verify the proposed method's ability to recognize teeth with high degree of accuracy. To test the feasibility and robustness of the model, a tenfold cross validation (CV) is presented in this paper. The result of tenfold CV effectively verifies the robustness of the model as the average F1 score obtained is more than 0.970. Thus, the proposed model can be used as a useful and reliable tool to assist dental care professionals in dentistry.

Bone Tunnel Placement Determination Method for 3D Images and Its Evaluation for Anterior Cruciate Ligament Reconstruction.

BACKGROUND: Anterior cruciate ligament (ACL) injury causes knee instability which affects sports activity involving cutting and twisting motions. The ACL reconstruction surgery replaces the damaged ACL with artificial one which is fixed to the bone tunnels opened by the surgeon. The outcome of the ACL reconstruction is strongly related to the placement of the bone tunnels, therefore, the optimization of tunnel drilling technique is an important factor to obtain satisfactory surgical results. AIMS: The quadrant method is used for the post-operative evaluation of the ACL reconstruction surgery, which evaluates the bone tunnel opening sites on the lateral 2D X-ray radiograph. METHODS: For the purpose of applying the quadrant method to the pre-operative knee MRI, we have synthesized the pseudo lateral 2D X-ray radiograph from the patients' knee MRI. This paper proposes a computer-aided surgical planning system for the ACL reconstruction. The proposed system estimates appropriate bone tunnel opening sites on the pseudo lateral 2D X-ray radiograph synthesized from the pre-operative knee MRI. RESULTS: In the experiment, the proposed method was applied to 98 subjects including subjects with osteoarthritis. The experimental results showed that the proposed method can estimate the bone tunnel opening sites accurately. The other experiment using 36 healthy patients showed that the proposed method is robust to the knee shape deformation caused by disease. CONCLUSION: It is verified that the proposed method can be applied to subjects with osteoarthritis.

Spatiotemporal Statistical Shape Model for Temporal Shape Change Analysis of Adult Brain.

BACKGROUND: This study presents a novel method of constructing a spatiotemporal statistical shape model (st-SSM) for adult brain. St-SSM is an extension of statistical shape model (SSM) in the temporal domain which will represent the statistical variability of shape as well as the temporal change of statistical variance with respect to time. AIMS: Expectation-Maximization (EM) based weighted principal component analysis (WPCA) using a temporal weight function is applied where the eigenvalues of each data are estimated by Estep using temporal eigenvectors, and M-step updates Eigenvectors in order to maximize the variance. Both E and M-step are iterated until updating vectors reaches the convergence point. A weight parameter for each subject is allocated in accordance with the subject's age to calculate the weighted variance. A Gaussian function is utilized to define the weight function. The center of the function is a time point while the variance is a predefined parameter. METHODS: The proposed method constructs adult brain st-SSM by changing the time point between minimum to maximum age range with a small interval. Here, the eigenvectors changes with aging. The feature vector of representing adult brain shape is extracted through a level set algorithm. To validate the method, this study employed 103 adult subjects (age: 22 to 93 y.o. with Mean ± SD = 59.32±16.89) from OASIS database. st-SSM was constructed for time point 40 to 90 with a step of 2. RESULTS: We calculated the temporal deformation change between two-time points and evaluated the corresponding difference to investigate the influence of analysis parameter. An application of the proposed model is also introduced which involves Alzheimer's disease (AD) identification utilizing support vector machine. CONCLUSION: In this study, st-SSM based adult brain shape feature extraction and classification techniques are introduced to classify between normal and AD subject as an application.

Stride-Phase Kinematic Parameters That Predict Peak Elbow Varus Torque.

BACKGROUND: During baseball pitching, a high amount of elbow varus torque in the arm cocking-to-acceleration phase is thought to be a biomechanical risk factor for medial elbow pain and injury. The biomechanics of the stride phase may provide preparation for the arm cocking-to-acceleration phase that follows it. PURPOSE: To determine the kinematic parameters that predict peak elbow varus torque during the stride phase of pitching. STUDY DESIGN: Descriptive laboratory study. METHODS: Participants were 107 high school baseball pitchers (age range, 15-18 years) without shoulder or elbow problems. Whole-body kinematics and kinetics during fastball pitching were analyzed using 3-dimensional measurements from 36 retroreflective markers. A total of 26 kinematic parameters of the upper and lower limbs during the stride phase leading up to the stride foot contact were extracted for multiple regression analysis to assess their combined effect on the magnitude of peak elbow varus torque. RESULTS: Increased wrist extension, elbow pronation, knee flexion on the leading leg, knee extension on the trailing leg at stride foot contact, and upward displacement of the body's center of mass in the stride phase were significantly correlated with decreased peak elbow varus torque (all P < .05). Moreover, 38% of the variance in peak elbow varus torque was explained by a combination of these 5 significant kinematic variables (P < .001). CONCLUSION: We found that 5 kinematic parameters during the stride phase and the combination of these parameters were associated with peak elbow varus torque. The stride phase provides biomechanical preparation for pitching and plays a key role in peak elbow varus torque in subsequent pitching phases. CLINICAL RELEVANCE: The present data can be used to screen pitching mechanics with motion capture assessment to reduce peak elbow varus torque. Decreased peak elbow varus torque is expected to reduce the risk of elbow medial pain and injury.

Intraoperative kinematic analysis of posterior stabilized total knee arthroplasty with asymmetric helical post-cam design.

PURPOSE: To investigate intraoperative kinematics during passive flexion using a surgical navigation system for knees undergoing posterior stabilized (PS) total knee arthroplasty (TKA) with an asymmetric helical post-cam design using navigation system. METHODS: In total, 45 knees with both pre- and postoperative kinematic data available were included in the study. Intraoperative kinematic measurements were performed during the course of surgery using the software incorporated in the navigation system. Measurements were performed at the following two time points: (1) before TKA procedure and (2) after TKA implantation. Among the kinematic parameters studied, anterior/posterior translation and axial rotation during flexion were subjected to the analysis. RESULTS: Before surgery, physiologic anterior/posterior translational pattern of the tibia during flexion (rollback of the femur) was found in only 15.6% of the knees. After TKA implantation, postoperative kinematic measurement showed no significant change in the tibial translational during knee flexion. Similarly, with regard to rotation, non-physiologic external tibial rotation in early flexion was observed in the majority of the knees before surgery, and this abnormal kinematic pattern remained after the TKA procedure. CONCLUSIONS: The intraoperative three-dimensional motion analysis using a navigation system showed that the physiologic kinematic pattern (anterior translation and internal rotation of the tibia during flexion) of the knee was distorted in osteoarthritic knees undergoing TKA. The abnormal kinematic pattern before surgery was not fully corrected even after implantation of the PS TKA designed to induce natural knee motion; however, no clear relationship between the intraoperative kinematic pattern and knee flexion angle at one year was demonstrated, and the effect of knee kinematics on postoperative knee function and patient's satisfaction is still unclear.

Gap Effect Abnormalities during a Visually Guided Pro-Saccade Task in Children with Attention Deficit Hyperactivity Disorder.

Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder that starts in early childhood and has a comprehensive impact on psychosocial activity and education as well as general health across the lifespan. Despite its prevalence, the current diagnostic criteria for ADHD are debated. Saccadic eye movements are easy to quantify and may be a quantitative biomarker for a wide variety of neurological and psychiatric disorders, including ADHD. The goal of this study was to examine whether children with ADHD exhibit abnormalities during a visually guided pro-saccadic eye-movement and to clarify the neurophysiological mechanisms associated with their behavioral impairments. Thirty-seven children with ADHD (aged 5-11 years) and 88 typically developing (TD) children (aged 5-11 years) were asked to perform a simple saccadic eye-movement task in which step and gap conditions were randomly interleaved. We evaluated the gap effect, which is the difference in the reaction time between the two conditions. Children with ADHD had a significantly longer reaction time than TD children (p < 0.01) and the gap effect was markedly attenuated (p < 0.01). These results suggest that the measurement of saccadic eye movements may provide a novel method for evaluating the behavioral symptoms and clinical features of ADHD, and that the gap effect is a potential biomarker for the diagnosis of ADHD in early childhood.

From cartoon to real time MRI: in vivo monitoring of phagocyte migration in mouse brain.

Recent studies have demonstrated that immune cells play an important role in the pathogenesis of many neurological conditions. Immune cells constantly survey the brain microvasculature for irregularities in levels of factors that signal homeostasis. Immune responses are initiated when necessary, resulting in mobilisation of the microglial cells resident in the central nervous system (CNS) and/or of infiltrating peripheral cells. However, little is known about the kinetics of immune cells in healthy and diseased CNS, because it is difficult to perform long-term visualisation of cell motility in live tissue with minimal invasion. Here, we describe highly sensitive in vivo MRI techniques for sequential monitoring of cell migration in the CNS at the single-cell level. We show that MRI combined with intravenous administration of super-paramagnetic particles of iron oxide (SPIO) can be used to monitor the transmigration of peripheral phagocytes into healthy or LPS-treated mouse brains. We also demonstrate dynamic cell migration in live animal brains with time-lapse MRI videos. Time-lapse MRI was used to visualise and track cells with low motility in a control mouse brain. High-sensitivity MRI cell tracking using SPIO offers new insights into immune cell kinetics in the brain and the mechanisms of CNS homeostasis.

Noninvasive cellular quantity measurement in Bone Marrow Stromal Cells/ Beta-tricalcium phosphate.

This paper describes noninvasive cellular quantity measurement in Bone Marrow Stromal Cells/ ß-tricalcium phosphate. We attempt to identify cellular quantity with an ultrasonic system. The ultrasonic waves are reflected at boundaries where there is a difference in acoustic impedances of the materials on each side of the boundary. Therefore, we focus on the reflected signal. From the obtained ultrasonic data, we extract two features; amplitude and frequency. Amplitude is obtained from the raw ultrasonic wave, and frequency is calculated from frequency spectrum obtained by applying cross-spectrum method. Therefore, we suggest the superiority of frequency to analyze Bone Marrow Stromal Cells. This study shows the ability of intervention to produce the desired beneficial effect.

Fuzzy connectedness image segmentation for newborn brain extraction in MR images.

Newborn's brain has a various shape, and easily changes with not only brain developing and cerebral diseases. Although the brain segmentation in MR images is an effective way to quantify the brain shape and size, there are few studies in neonatal brain MR image analysis. This paper introduces a novel method based on fuzzy connectedness (FC) with fuzzy object model (FOM). FOM is built from a training dataset, and gives fuzzy degree belonging to parenchyma with respect to location and intensity. FC is calculated from object affinity and homogeneous affinity, and the object affinity is given by the FOM. The method first segments the white matter, and then segments the surrounding cortex. The propose method has been applied to 10 newborn subjects whose revised age was between -1 month and +2 month. Leave-on-out cross-validation (LOOCV) was conducted, and the mean false-positive volume fraction was 1.33%, the mean false-negative volume fraction was 2.90%, and geometric-mean was 1.42%.

Automated pelvic anatomical coordinate system is reproducible for determination of anterior pelvic plane.

Most of computer-assisted planning systems need to determine the anatomical axis based on the anterior pelvic plane (APP). We analysed that our new system is more reproducible for determination of APP than previous methods. A pelvic model bone and two subjects suffering from hip osteoarthritis were evaluated. Multidetector-row computed tomography (MDCT) images were scanned with various rotations by MDCT scanner. The pelvic rotation was calibrated using silhouette images. APP was determined by an optimisation technique. The values of variation of APP caused by pelvic rotation were analysed with statistical analysis. APP determination with calibration and optimisation was most reproducible.The values of variance of APP were within 0.05° in model bone and 0.2° even in patient pelvis. Furthermore, the values of variance of APP with calibration/optimisation were significantly lower in comparison without calibration/optimisation. Both calibration and optimisation are actually required for determination of APP. This system could contribute to the evaluation of hip joint kinematics and computer-assisted surgery.

A neonatal brain MR image template of 1 week newborn.

PURPOSE: Magnetic resonance imaging (MRI) is often used to detect and treat neonatal cerebral disorders. However, neonatal MR image interpretation is limited by intra- and inter-observer variability. To reduce such variability, a template-based computer-aided diagnosis system is being developed, and several methods for creating templates were evaluated. METHOD: Spatial normalization for each individual's MR images is used to accommodate the individual variation in brain shape. Because the conventional normalization uses as adult brain template, it can be difficult to analyze the neonatal brain, as there are large difference between the adult brain and the neonatal brain. This article investigates three approaches for defining a neonatal template for 1-week-old newborns for diagnosing neonatal cerebral disorders. The first approach uses an individual neonatal head as the template. The second approach applies skull stripping to the first approach, and the third approach produces a template by averaging brain MR images of 7 neonates. To validate the approaches, the normalization accuracy was evaluated using mutual information and anatomical landmarks. RESULTS: The experimental results of 7 neonates (revised age 5.6 ± 17.6 days) showed that normalization accuracy was significantly higher with the third approach than with the conventional adult template and the other two approaches (P < 0.01). CONCLUSION: Three approaches to neonatal brain template matching for spinal normalization of MRI scans were applied, demonstrating that a population average gave the best results.